Abstract | ||
---|---|---|
Most Wi-Fi based localization algorithms are cooperative as user device is required to associate with an AP. However, user may not associate with AP in scenarios such as supermarkets which calls for non-cooperative localization. In this paper, the probe request (PR) frame sent by device is analyzed and the weighted kernel density estimation assisted Bayes (w-KAB) algorithm is utilized for localization. The PR frame is sent in a sparse manner in time and the probability distribution of its receiving signal strength is complicated due to channel misalignment. Therefore kernel density estimation is adopted in the training stage to estimate the distribution of signal strength accurately with a limited amount of training data. In the localization stage, a weighted naive Bayes algorithm is used to estimate the location of user. Experiments are also conducted using off the shelf devices to validate the performance of the proposed algorithm. |
Year | Venue | Keywords |
---|---|---|
2016 | 2016 IEEE 84TH VEHICULAR TECHNOLOGY CONFERENCE (VTC FALL) | Non-cooperative localization, Kernel density estimation, Naive Bayes |
Field | DocType | ISSN |
Kernel (linear algebra),Naive Bayes classifier,Computer science,Communication channel,Computer network,Algorithm,Real-time computing,Probability distribution,Probability density function,Variable kernel density estimation,Kernel density estimation,Bayes' theorem | Conference | 2577-2465 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hao Chen | 1 | 156 | 61.18 |
Yifan Zhang | 2 | 30 | 10.85 |
Wei Li | 3 | 422 | 56.67 |
zhang | 4 | 210 | 25.85 |